EPIDEMIOLOGY
Acts of sexual aggression have become increasingly commonplace. There has been an increase in both violent crimes committed by juveniles (Office of Juvenile Justice and Delinquency Prevention, 1994a) and in the reports of sexual aggression and sexual abuse (Hampton, 1995). The National Committee to Prevent Child Abuse (1998) estimates that of the 1 million cases of confirmed child maltreatment in 1997, approximately 8% were the victims of sexual abuse. Sexual assault is one of the fastest-growing violent crimes in the United States and accounts for 7% of all violent crimes (Hampton, 1995).
While adolescents (15-18 years) make up only 6% of the population of the United States, they commit 25% of the "index crimes," such as arson, homicide, manslaughter, robbery, aggravated assault, burglary, larceny, and forcible rape (Siegel and Senna, 1988). From 1983 to 1992 there was a reported 8% to 10% increase in the proportion of adolescents involved in some type of serious violent offense (Elliott, 1994a). Forcible rape arrests by juveniles increased 20% between 1983 and 1992 (Office of Juvenile Justice and Delinquency Prevention, 1994a). Violent behavior is highly correlated with being male at a specific phase of life. The highest risk for the initiation of serious violent behavior occurs between 15 and 16 years (Elliott, 1994a). In Elliott's (1994a, b) longitudinal study of a national probability sample of 1,725 youths (11-17 years), he found that the prevalence of serious violent offenses such as rape that involved some injury or use of a weapon peaked at 17 years of age for males and 15 to 16 years for females. Three gender differences were apparent:
* Violent behavior occurs earlier in females.
* The subsequent decline in violent behavior is steeper for females.
* The gender differences become greater over time.
The increase in the reports of youthful sexual aggression has been well documented (Elliott, 1994a--c; Office of Juvenile Justice and Delinquency Prevention, 1994b). Ageton (1983) concluded from a probability sample of 863 male adolescents 13 to 19 years of age that the rate of sexual assault per 100,000 adolescent males ranged from 5,000 to 16,000. survey of high school students revealed that 1 out of 5 had been involved in forcing sex on another and that 60% of the boys found it acceptable in one or more situations for a boy to force sex on a girl (Davis et al., 1993). In their national survey of 1,600 sexually abusive youths, Ryan et al. (1996) reported that they came from all racial, economic, ethnic, and religious backgrounds and generally mirrored the population. They ranged in age from 5 to 21 years, with a model age of 14 years, and were predominantly male. While the majority of juvenile sexual abusers are adolescents, there is increasing concern about the sexually aggressive and exploitative behavio rs in prepubescent and latency-age children (Araji, 1997; Gil and Johnson, 1993; Johnson, 1988).
THE SPECTRUM OF SEXUAL OFFENSES
The spectrum of sexually inappropriate behaviors ranges from various forms of sexual harassment and nonconract sexual behaviors such as obscene phone calls, exhibitionism, and voyeurism to varying degrees of sexual aggression involving direct sexual contact including frottage, fondling, digital and penile penetration, fellatio, sodomy, and other aggressive sexual acts.
A form of sexual aggression which is becoming more visible is child-on-child sexual harassment, which is unwanted sexual attention being directed to other children and adolescents. Child-on-child sexual harassment has become increasingly evident in our schools. Straus (1988) found that approximately 50% of high school junior and senior girls reported incidents of sexual harassment. Ponton (1996) reported that sexual harassment is found throughout the school environment and that student-to-student harassment is more common than adult-to-student harassment. It is known that child victims of harassment experience an array of emotional and behavioral effects (Ponton, 1996).
The spectrum of sexual offending behavior reported in the literature varies depending on the sample studied. For example, children and adolescents sampled in detention centers, residential treatment programs, and outpatient clinics report different spectra of sexually offensive behavior.
Ryan et al. (1996) found in a survey of sexually abusive youths from diverse outpatient and residential programs that they had participated in a wide range of sexual offenses. Seventy percent of the sexual offenses involved penetration and/or oral genital behavior: vaginal or anal penetration without oral-genital contact, 35%; oral-genital contact, 14.7; penetration and oral-genital contact, 18%.
Studies of outpatient populations of juvenile sexual abusers indicate that the most common sexual offenses are fondling or indecent liberties," 40% to 60%; rape and/or sodomy, 20% to 40%; and nonconract sexual offenses, 5% to 10% (Fehrenbach et al., 1986; Smith and Monastersky 1986). The average juvenile sex offender younger than 18 years of age has committed 8 to 9 sexual offenses and averaged 4 to 7 victims (Abel et al., 1986; Shaw et al., 1993).
While adolescents usually use coercion in the process of committing sexual offenses, they are less likely to physically harm their victims than are adult sex offenders. The coercion is usually expressed as bribery, intimidation, threats of harm or violent injury, physical force, and rarely the use of a weapon (Fehrenbach et al., 1986; Knight and Prentky, 1993). Nevertheless, it is well known that victims report higher levels of coercion and force than is self-reported by offenders (Davis and Leitenberg, 1987). Fehrenbach et al. (1986) found that 22% of the offenders continued their aggressive sexual acts even when the victims expressed "hurt or fear."
VICTIM CHARACTERISTICS
The great majority of the victims of male sex offenders are female (Davis and Leitenberg, 1987; Fehrenbach et al., 1986; Longo, 1982; Ryan et al., 1996). Adolescent sex offenders commit most of their sexual assaults against boys (Hunter and Becker, 1999). Ninety percent of sexual abuse victims are between 3 and 16 years of age. The majority of the victims of juvenile sex offenders are younger than 9 years of age, and approximately 25% to 40% are younger than 6 years of age (Abel and Osborn, 1992; Fehrenbach et al., 1986; Ryan et al., 1996). Boys who are victimized tend to be younger than their female counterparts (Hunter, 1991).
Distinctions have been made between juveniles who commit incest and those who select victims outside of the family. Approximately 40% of victims are relatives. Worling (1995) compared adolescent sex offenders who victimized siblings and those who selected victims outside of their family. He found that those who committed incest came from families with significantly more marital discord, parental rejection, physical discipline, and negative family atmosphere and were generally more dissatisfied with their family relationships. These offenders were more likely to have been sexually victimized.
It has generally been believed that incest offenders are exclusively incestuous and thus more treatable (Becker, 1994; Becker et al., 1993). Abel et al. (1988), however, studied 159 adult incest offenders and found their sexual offenses frequently extended beyond the family. Forty-nine percent had histories of being involved in nonincestual female pedophilia, 12% in male pedophilia, and 19% in rape.
THE CYCLE OF VIOLENCE
The cycle of violence is passed from one generation to the next. Early childhood victimization increases the risk for subsequent delinquent behavior, violent behavior, and criminality (Hunter and Becker, 1999; Kobayashi et al., 1995; Widom, 1989). There is often a significant history of child maltreatment including neglect and physical and sexual abuse in the early lives of juvenile sex offenders (Knight and Prentky, 1993; Ryan et al., 1996). Kobayashi et al. (1995) found that physical abuse by fathers and sexual abuse by males increased sexual aggression by adolescents. While the exposure of juvenile sex offenders to physical violence is comparable with that generally found in other delinquents, emphasis has been placed on the increased reports of sexual victimization in the history of sexual abusers.
Sexual behavior is commonly affected by a history of sexual victimization. Boys and girls exposed to sexual abuse and deviant sexual experiences, behaviors, attitudes, and knowledge are at risk for early erotization and precocious sexualization. There is some evidence that boys are more likely than girls to be sexually aroused at the time of sexual victimization and to subsequently exhibit more sexual behaviors (Friedrich, 1995; McClellan et al., 1997).
A history of sexual abuse is more prevalent in sexual abusers than in the general population and in nonsexual abusers (Dhawan and Marshall, 1996; Friedrich, 1995). Reports of sexual victimization in the history of adolescent sex offenders vary from 19% to 82% (Becker et al., 1986a; Dhawan and Marshall, 1996; Fehrenbach et al., 1986; Longo, 1982; Kahn and Chambers, 1991; Ryan et al., 1996; Shaw et al., 1993; Zgourides et al., 1994). The victims of sexual abuse may internalize the aggressive and erotized facets of their sexual experiences into preferred patterns of deviant sexual gratification through a process of social learning, imitation, modeling, and identificatory pathways. Becker et al. (1989, 1992) demonstrated that a history of physical and sexual abuse in juvenile sex offenders is associated with higher phallometrically measured arousal to both deviant and nondeviant sexual stimuli. They found that adolescents who had been sexually victimized manifested more deviant erectile responses than adolescent s who had not been sexually abused. Boys who have been sexually abused often demonstrate higher rates of externalized behaviors such as aggression, impulsivity, and temper tantrums as well as sexual preoccupations and sexually inappropriate behaviors (Holmes and Slap, 1998). Cooper et al. (1996) compared sexually abused and nonabused adolescent sexual abusers and found that the abused sexual abuser had an earlier onset of sexual offending behavior, had more victims, and was more likely to have both male and female victims and to manifest greater psychopathology and interpersonal problems.
Early sexual sensitizing experiences may determine deviant patterns of erotization. Developmental experiences with significant others are crucial to the elaboration of a sexual life. It is known that animate or inanimate non-sexually arousing stimuli may through pairing with sexually arousing stimuli acquire a conditioned sexually evocative status. Patterns of sexual arousal may become coincidentally paired with aggressive sexuality, enemas, corporal punishment, abusive discipline, or other accidental and sometimes terrifying experiences. These early experiences, if associated with sexual arousal and sexual excitement, may become woven into the preferred patterns of sexual gratification. Boys and girls who have been sexually victimized may experience their first sexual arousal during the victimization. The factors related to and derived from the sexual abuse experience which are thought to be associated with increased risk for inappropriate sexual behaviors are sexual arousal at the time of the sexual abuse, uncertainty and confusion about sexual identity, compensatory hypermasculinity, and a readiness to reenact the sexually victimizing experience (Friedrich, 1995; Watkins and Bentovim, 1992).
There is increasing awareness of the importance of early sexual abuse experience in precipitating inappropriate sexual behaviors in very young children (Holmes and Slap, 1998). In a retrospective chart review of 499 mentally ill youths (5-18 years), McClellan et al. (1996) found that the probability of engaging in any sexually inappropriate behavior was inversely related to the age at which they were sexually abused. A history of sexual abuse before 7 years of age was significantly associated with hypersexuality, exposing, and victimizing sexual behaviors. A number of studies have found that sexually abused children may become sexually reactive and are more sexually inappropriate and manifest higher ratings on sexual activities scales than nonabused children (Deblinger et al., 1989; Friedrich, 1993). Yates (1982) described sexually victimized children, aged 2 to 6 years, who were found to exhibit excessive erotization. These children had trouble differentiating sensual from affectionate touching and erotic f rom nonerotic relationships. They were easily sexually aroused and quick to turn to others for sexual gratification. Kendall-Tackett et al. (1993, p. 165) noted that children who had been sexually abused were more likely to exhibit sexualized behaviors such as "sexualized play with dolls, putting objects into the anuses or vaginas, excessive or public masturbation, seductive behavior, requesting sexual stimulation and age inappropriate sexual knowledge."
It is known that the younger the child when his or her first sexual offense is committed, the more likely that child has been sexually victimized (Johnson, 1988; McClellan et al., 1996). Johnson (1988) found that 72% of child sex offenders younger than 6 years of age, 42% of those 7 to 10 years old, and 35% of those 11 to 12 years old had been sexually victimized. Shaw et al. (1993, 1996) documented the high prevalence of early sexual victimization in the history of adolescent sex offenders (67%) who had been committed to a residential center. They found that the mean age of sexual victimization was 6 years, with 35% of the group having been sexually abused before age 5 years. While 30% of the adolescent sex offenders were not directly sexually abused, they invariably came from homes where they were prematurely exposed to sexual violence, promiscuity, and pornography. Ford and Linney (1995) observed that juvenile sex offenders had been exposed to pornographic materials and/or "hard core" sex magazines at an earlier age than non-sexually violent offenders. Longo (1982) found that among the adolescent sex offenders he studied there was a higher than expected history of direct, consenting sexual contact with older males and females. These adolescents reported feeling uncomfortable, insecure, inexperienced, inferior, and unable to satisfy their older sex partners.
As important as sexual victimization is in the history of juvenile sexual abusers, it is not explanatory. Most victims of sexual abuse do not grow up to be abusers. As a form of child maltreatment it is only one of a number of critical factors that may contribute to the risk of becoming sexually abusive (Prentky et al., 1997). Nevertheless it is essential to understand the role of sexual victimization in the patterning of sexual offending behavior and as a critical factor in the planning of therapeutic interventions.
CLINICAL PRESENTATION
Early studies of inappropriate sexual behavior in juveniles were concerned with trying to define a child molester syndrome or profile (Shoor et al., 1966). The complex, multidetermined nature of juvenile sexually aggressive behavior has made it difficult to set up a predictable taxonomic system. At present, there is no evidence that any one profile or typology is characteristic of juvenile sex offenders (Becker and Hunter, 1993; Levin and Stava, 1987).
Adolescents who display sexually abusive behavior are a heterogeneous population. They are represented in every socioeconomic, racial, ethnic, religious, and cultural group. There are, however, a number of factors that have been commonly found in the history of adolescent sex offenders. These include impaired social and interpersonal skills, prior delinquent behavior, impulsivity, academic and school problems, family instability, family violence, abuse and neglect, and psychopathology (Araji, 1997; Becker and Hunter, 1993; Fehrenbach et al., 1986; Ryan et al., 1996).
THE PREADOLESCENT SEX ABUSER
There is increasing awareness that children younger than 12 years of age may be sexually abusive toward other children (Araji, 1997; Gray et al., 1997; Johnson, 1988, 1989, 1999). In these cases it is necessary to discriminate between age-appropriate sexual exploration and sexually abusive behavior. Appropriate sexual exploratory behavior is usually carried out with children of the same age and size and with mutual consent.
The age differential between the abuser and the victim in children is less definitive. It may be less than 2 years, and in some instances the younger child may be the initiator of sexual behaviors. For children aged 9 years and older, the age difference between abuser and victim is usually greater than 2 years.
Younger sexual abusers are often reacting to their own histories of sexual victimization (Gil and Johnson, 1993; Johnson, 1988). The term sexually reactive refers to children who display sexually inappropriate behavior in response to sexual abuse or exposure to explicit sexual stimuli (Gil and Johnson, 1993). They have been characterized as often exhibiting impulsivity, anger, fear, loneliness, confusion, depression, obsessional and compulsive preoccupation with sex, excessive sexualization, anxiety, and sleep disturbance (Araji, 1997; Johnson and Berry, 1989). In approximately half of the cases, the sexual victims of these children are their own siblings. The sexually abusive behaviors between children are usually characterized by themes of secrecy, dominance, coercion, threat, and force. The sexual behaviors are advanced far beyond those expected for the age of the child and may include oral and vaginal intercourse and forcible penetration of the anus or vagina with fingers or other objects. The sexual beh aviors are patterned, increase over time, are often associated with conduct-disordered behaviors, and may continue in spite of intervention (Araji, 1997; Gil and Johnson, 1993).
A study of 6- to 12-year-old children who had exhibited sexual misbehaviors found a number of characteristics indicative of parental and family distress, i.e., violence between parents, history of sexual victimization and perpetration within the extended family, physical abuse, poverty, the witnessing of domestic violence, parental arrests, and the prior use of educational and therapeutic services for children (Gray et al., 1997). Younger children exhibiting sexual misbehaviors are more likely to have been sexually and physically abused at a younger age and to manifest more problematic sexual behaviors, with a higher percentage of hands-on sexual behaviors.
THE ADOLESCENT SEX ABUSER
Becker (1988) suggested that there are essentially 4 kinds of sexual abusers, with most offenders combining features of each:
* The true paraphiliac with a well-established deviant pattern of sexual arousal.
* The antisocial youth whose sexual offending behavior is but one facet of his or her opportunistically exploiting others.
* The adolescent compromised by a psychiatric or neurological! Biological substrate disorder which interferes with his or her ability to regulate and modulate aggressive and sexual impulses.
* The youth whose impaired social and interpersonal skills result in turning to younger children for sexual gratification unavailable from peer groups.
Sexually aggressive adolescents are more likely to endorse traditional sex-role stereotypes, male dominance, and rape-supportive myths and to have negative and stereotypical attitudes about women (Epps et al., 1993; Segal and Stermac, 1984; White and Koss, 1993). The history of prior sexual offenses at the time of apprehension varies from 8% to 65% depending on whether the history was obtained from self-report or prior offense records and the specific sample of sexual abusers studied (Fehrenbach et al, 1986; Ryan et al., 1996; Shaw et at., 1996; Smith and Monastersky, 1986).
PSYCHOPATHOLOGY
Sexual offending behavior is associated with a matrix of behavioral, emotional, and developmental problems (Awad and Saunders, 1989; Kavoussi et at., 1988; Shaw et al., 1993, 1996). The studies of the psychological characteristics of sexual abusers, however, may be skewed as only a small percentage of sexual abusers are actually caught (Harvey and Herman, 1992).
Investigators have examined adolescent sexual abusers in an effort to explicate the spectrum of psychopathology (Becker, 1988; Kavoussi et al., 1988; Lewis et al., 1979; Shaw et al., 1993, 1996). The findings are understandably affected by the assessment procedures, diagnostic instruments, and sample selection, i.e., inpatient-outpatient and severity of offending behavior.
When adolescent sexual abusers are compared with psychiatric controls or normal subjects on the MMPI, they manifest more psychopathology, demonstrate more global maladjustment, and are more threatened by heterosexual interactions (Herkov et al., 1996; Katz, 1990).
Studies comparing violent and nonviolent delinquents with juvenile sexual abusers across a broad range of intellectual, psychological, neuropsychological, and psychoeducational variables have found few significant differences between the groups (Hastings et al., 1997; Jacobs et al., 1997; Lewis et al., 1979; Tarter et al., 1983). Truscotr (1993) compared sexual, violent, and property offenders and noted that the sex offenders were twice as likely to have a history of being sexually abused. Juvenile sex offenders generally do less well on academic achievement tests than nonsexual delinquents and youths with conduct disorder (Lewis et al., 1979; Shaw et al., 1993).
A number of studies have focused on the specific mental disorders found among juvenile sex offenders. Psychiatric comorbidity has been found in approximately 60% to 90% of adolescent sexual abusers. The most prevalent comorbid psychiatric disorders are conduct disorder, 45% to 80%; mood disorders, 35% to 50%; anxiety disorders, 30% to 50%; substance abuse, 20% to 30%; and attention-deficit hyperactivity disorder, 10% to 20% (Becker et al., 1986b, 1991; Kavoussi et al., 1988; Shaw et al., 1993). The younger the child when his or her first sexual offense was committed, the higher the number of coexisting psychiatric diagnoses (Shaw et al., 1996).
DELINQUENCY
Reports of prior nonsexual delinquent behavior in the history of juvenile sexual abusers vary from 27% to 91% (Awad and Saunders, 1989; Becker et al., 1986a; Fehrenbach et al., 1986; Ryan et al., 1996; Saunders et al., 1986; Shoot et al., 1966; Shaw et al., 1993). A national survey of juvenile sex offenders reported that 63% had previously committed non-sexual delinquent offenses (Ryan et al., 1996). Twenty-eight percent had a history of committing 3 or more nonsexual criminal offenses.
Elliott (1994b), in the National Youth Survey, a longitudinal study of a national probability sample of 1,725 youths aged 11 to 17 years, reported that rape is the final step in an escalating sequence of violent criminal activity. The usual sequence is of criminal behavior evolving from aggravated assault to robbery to rape. The peak age for males to commit a sexual assault was 17 years. Rapists were noted to have committed virtually every form of violent offense. It is evident that there is a subset of sexually aggressive offenders characterized by core antisocial features in which the sexual aggression is only one facet of a lifestyle in which the individual opportunistically exploits others for personal gain (Ageton, 1983; Hastings et al., 1997; Lewis et al., 1979; Shaw et al., 1993).
PERSONALITY CHARACTERISTICS
Adult sex offenders frequently manifest significant personality disturbances (Packard and Rosner, 1985). In an effort to explicate personality traits and character pathology, Shaw et al. (1996), using a structured psychiatric interview, examined a population of adolescent sex offenders in a residential center. They found a high prevalence of severe personality traits including narcissistic, borderline, and conduct-disordered behaviors. Ninety-two percent of the boys met criteria for a diagnosis of conduct disorder; 67%, narcissistic personality disorder; and 72%, borderline personality disorder. The younger the age of onset of sexual offending behavior and the younger the sex offender was at the time of his or her own sexual victimization, the more likely he or she was to exhibit symptoms of a borderline personality. The high prevalence of narcissistic and borderline psychopathology is consistent with histories of severe emotional, physical, and sexual abuse. Carpenter et al. (1995) found that adolescent sex abusers who offended against children were more schizoid, dependent, and avoidant than adolescents who sexually abused peers.
FAMILY ENVIRONMENT
Most juvenile sex offenders live at home at the time of committing their sexual offenses (Kahn and Chambers, 1991; Ryan et al., 1996). The family environment is usually characterized by family conflict, family instability, family dysfunction, and exposure to violence (Awad et al., 1984; Becker et al., 1993); harsh, inconsistent parenting and physical and sexual maltreatment (Becker et al., 1993; Knight and Prentky, 1993; Lewis et al., 1979; Shaw et al., 1993); and low levels of adaptability and cohesion (Smith and Monastersky, 1986).
SOCIAL AND INTERPERSONAL SKILLS
Adolescent sex offenders manifest impaired social and interpersonal skills. Fehrenbach and Monastersky (1988) found that 65% were socially isolated and that 32% did not have a friend. Awad et al. (1984) reported that 46% of their sample of adolescent sex offenders were loners. There is some evidence that adolescent sex offenders unable to relate to their own peer group frequently turn to younger children for the gratifications denied to them by their own peer group (Awad and Saunders, 1989; Shoor et al., 1966).
SCHOOL AND ACADEMIC PROBLEMS
Juvenile sex offenders usually present with a history of academic and school behavior problems and perform less well on tests of academic skills (Awad et al., 1984; Awad and Saunders, 1989; Fehrenbach and Monastersky, 1988; Lewis et al., 1979; Ryan et al., 1996; Shaw et al., 1993). Studies have found that approximately 40% to 80% of juvenile sex offenders manifest learning disabilities and behavior problems in school (Awad and Saunders, 1989; Ryan et al., 1996; Shaw et al., 1993).
THE FEMALE JUVENILE SEX ABUSER
There have been few studies of female sexual aggression (Fehrenbach and Monastersky 1988; Green, 1999; Green and Kaplan, 1994; Kaplan and Green, 1995; Knopp and Lackey, 1987; Mathews et al., 1997). Anderson (1996) surveyed young college women and found that 29% had engaged in sexual coercion, 28% had committed sexual abuse, and 43% reported initiating sexual contact by using sexually aggressive strategies. It is known, however, that 50% to 95% of female sexual abusers report a history of sexual victimization (Green, 1999; Kaplan and Green, 1995). Female juvenile offenders are more likely to have been sexually abused at a younger age and to have had multiple abusers and are 3 times more likely to have been sexually abused by a female (Mathews et al., 1997).
Fehrenbach and Monstersky (1988) studied 28 female adolescent sex offenders, 10 to 18 years of age, with a mean age of 13.6 years. Fifteen had been charged with rape and 13 with raking indecent liberties. Ten of the female sexual abusers had assaulted males, 16 had assaulted females, and 2 had assaulted both males and females. Approximately 70% of the offenses took place while the offender was baby-sitting. The mean age of the victims was 5.2 years. Fourteen (50%) of the offenders had a history of being sexually abused.
Mathews et al. (1989) described 5 variations commonly found among adult female sex abusers: (1) the female adolescent who opportunistically exploits and fondles younger children; (2) the female with a history of being sexually abused who subsequently abuses younger children; (3) the older female who falls in love with a younger child whom she is mentoring; (4) the emotionally unstable female with an impaired capacity to modulate and regulate her own sexual impulses; and (5) the female who sexually abuses children only in the company of a male sexual abuser.
THE DEVELOPMENTALLY DISABLED JUVENILE SEX ABUSER
There is some suggestion that developmental disabilities are overrepresented in juvenile sex offenders. The prevalence of sexual offending behavior is at least as common if not more common in the disabled population (Gilby et al., 1989; Stermac and Sheridan, 1993). In a retrospective chart review of 200 seriously mentally ill youths (aged 5-18 years), McCurry et al. (1998) found that 50% had engaged in problem sexual behaviors and that independent of a history of sexual abuse, low verbal IQ was correlated with sexually inappropriate behavior. Twenty (65%) of those with Full Scale IQs (WISC-R/WAIS-R) less than 70 engaged in inappropriate sexual behaviors.
The clinical characteristics of disabled sex offenders, the spectrum of sexual offenses, and their victim profiles are not demonstrably different from those of nondisabled sex offenders (Stermac and Sheridan, 1993). The paucity of research studies limits our overall understanding of this group of sexual abusers. There is recognition that treatment strategies have to be tailored to their cognitive limitations and are usually more focused on behavioral interventions (Murphy et al., 1983).
ASSESSMENT
The assessment of sexually abusive young people is carried out for the purpose of understanding the sexual abuser so that one may make reasonable judgments regarding the protection of the community, case disposition, and treatment planning (Dougher, 1988a; Ross and Loss, 1991). The clinical assessment of juvenile sex abusers requires the same comprehensive evaluation as that of other children and adolescents (American Academy of Child and Adolescent Psychiatry, 1997). It necessarily requires the use of multiple informants and multiple sources of data. Important sources of information are medical and psychological reports, offense reports, victim statements, protective services reports, and probation reports. Ideally the collateral information should be obtained before the individual interview; otherwise, one is left relatively helpless before the offender's normal proclivity to minimize and deny. There are no empirical measures or psychological batteries that can substitute for a careful assessment.
FORENSIC CONSIDERATIONS
It is essential that the clinician define his or her role in the assessment of the sexual abuser. While many of the issues are relevant to a forensic evaluation, these parameters are designed to provide guidelines for the clinical evaluation of the sexual abuser. It is generally preferable to conduct the clinical evaluation after adjudication. The focus of the clinical interview is on assessing amenability to treatment, required levels of care, treatment goals and objectives, and risk of reoffending rather than a determination of guilt or innocence (Hunter and Lexier, 1999). The juvenile sex abuser is advised of reporting laws and the limits of professional confidentiality. An informed consent signed by the juvenile and the juvenile's parent/guardian should be obtained before the clinical interview. Consent forms should be developed to cover the use of controversial assessment and treatment procedures such as phallometric assessment, aversive conditioning, and medications that are not accepted as standard o f practice. The role of the protective services and the juvenile justice system may have to be explained. It is important to educate and clarify for the individual and the family what is going to happen and when.
One of the difficulties in evaluating sexually offending youths is the contradictory intentions of those who provide clinical services and representatives of the court who are interested in prosecution or defense. Information obtained by the clinician during the clinical interview may be subpoenaed by the court. The clinician may have to testify under oath as to what he or she knows about the sexual history and activities of the offender. In some instances, because of impending legal and court procedures, the attorney will tell the youth not to report to the clinician any sexual history or activities. In these instances only a partial and limited sexual history will be obtained, and it will be necessary to wait until the court has determined responsibility before completing the evaluation.
THE CLINICAL INTERVIEW
The cornerstone of the assessment and evaluation of the sexual abuser is an extensive and comprehensive individual clinical interview. It is common for sexual abusers to deny, minimize, and disavow their sexual offenses. Sefarbi (1990) found that half of adolescent sexual abusers initially denied the sexual offending behavior at the time of clinical referral and that this denial was usually supported by the family. In a national survey of juvenile sex offenders, Ryan et al. (1996) found that at the time of the evaluation only 19.4% accepted full responsibility for the sexual offense. Thirty-three percent admitted little or no responsibility. Sixty-two percent expressed little or no empathy for the victim. Fifty-one percent expressed little or no remorse. Although nearly two thirds accepted blame for the sexual offense, one third blamed the victim.
The denials usually take the form of protestations of innocence, statements that they are wrongly accused, or overt attempts to discount the victim's credibility. Not uncommonly, the offender states the victim consented to the sexual activity in spite of all the evidence to the contrary. Considerable time and effort will be necessary to unravel the underlying motivations and sexual fantasies.
It is necessary to establish the nature of the sexually abusive behavior. This is often a difficult task because of the interplay of a number of variables which may impede the recovery of the exact details of what transpired. Because in many cases laws have been transgressed, the offender is often less than forth-coming. Issues of shame, guilt, and fear of punishment impede disclosure. Most offenders are not motivated to disclose the circumstances of their sexually abusive behavior until they are confronted or fear the consequences if they do not talk. It is essential that the clinician adopt a nonjudgmental stance and relate to the juvenile offender in a matter-of-fact, exploratory manner. While the clinician has a structured agenda, it is advisable to follow the patient's lead and rake advantage of openended questions if there are promising opportunities. The clinician avoids sexual jargon, maintains a sense of seriousness, and avoids any minimization of the seriousness of the sexually abusive behavior. Ge nerally it is best to initiate the interview with a nonthreatening line of questioning and thereby avoid the initial defensiveness. The interviewer develops lines of questioning that will allow him or her to know more about the offender and the offender's family, school, and current life situation. The interviewer confronts minimization, denial, and the omission of important information. There is little value in getting angry and accusatory. It is more useful to be patient, persistent, and nor easily dissuaded.
Assessment of the Sexually Abusive Behavior
In the evaluation the clinician is interested in answering a number of questions regarding the abuser and the sexual abuse incident (Table 2) (Groth et al., 1981; Ross and Loss, 1991; Saunders and Awad, 1988).
Sexual History
As part of the ongoing psychiatric assessment it is essential to obtain a comprehensive sexual history. This is best accomplished after there has been some time to develop a trusting relationship. It works best to have a conceptual framework and sequence of questions which directs the inquiry.
In the evaluation, the interviewer determines the juvenile's sexual knowledge and education, sexual development, and sexual experiences. Inquiries are directed to learn what the juvenile knows about gender differences, sexual intercourse, autoerotic practices, and his or her preferred patterns of sexual behaviors. Specific questions may be asked about the juvenile's understanding and knowledge of normal sexual activities such as kissing, dating, petting, and masturbation and whether the juvenile has been sexually active and engaged in intercourse, homoerotic experiences, etc. It is necessary to discriminate between compulsive sexual behaviors and paraphiliac compulsive sexual behaviors. In the former condition the individual displays conventional and normative sexual behaviors, but they are excessive and carried to the extreme (Coleman, 1990, 1992). It has been suggested that nonparaphiliac compulsive sexual behaviors may be a variant of obsessive-compulsive disorder.
The clinician will need to explore the history of aggressive sexual behaviors including previous incidents of sexually abusive behavior; the pattern and spectrum of previously committed aggressive sexual acts; the victim profile; the internal and external triggers that initiate the sexual abuse cycle; the role of aggression and sadism in the sexual offense; the need to dominate, control, and humiliate the victim; the erotization of the aggression; the history of sexual victimization, physical abuse, and neglect; the history of exposure to inappropriate and sexually explicit behavior; and the history of prior nonsexual delinquent behavior.
Developmental and Psychosocial History
Other areas of the assessment process are those associated with a comprehensive developmental history, i.e., the nature of the pregnancy, perinaral history, developmental milestones, family relationships, early identificatory models, capacity for relationships, school experiences, social skills, substance abuse, and prior medical and psychiatric history. The family assessment provides an opportunity to understand the early developmental and environmental context within which the sexual abuser developed. Information is obtained regarding the parents' personal and psychological history, their use of authority and discipline, and the role of coercive sexuality in the family. How is affection, tenderness, competition, aggression, love, sexuality, and lust expressed in the family? How supportive and available is the family as a treatment resource?
Legal History
The clinician must ascertain whether there is a history of arrests, convictions, incarcerations, use of weapons, or cruelty to animals.
Medical and Psychiatric History
It is important to obtain a comprehensive medical and psychiatric history with specific attention to sexually transmitted diseases, human immunodeficiency virus infection, psychopathology, and psychiatric comorbidity.
School and Academic History
A specific area of concern is the evaluation of intellectual capacities and academic performance. Information is obtained from the school and from formal psychoeducarional and psychological assessments.
Mental Status Examination
A comprehensive mental status examination is carried out to assess the presence of psychopathology, personality disturbances, organicity, and substance abuse and to acquire an understanding of adaptive, coping, and defensive strategies. Suicidal content and risk should be assessed specifically. Apprehension by judicial authority and the associated shame of exposure, embarrassment, stigmatization, fear of punishment, and incarceration are risk factors for suicidal behavior. A careful assessment of the spectrum of suicidal behavior is undertaken to establish the degree to which suicidal ideation and history of suicidal behaviors, threats, or plans are present.
PSYCHOLOGICAL TESTING
There are no specific empirical measures or psychometric tests which can identify diagnose, or classify sexual abusers. Prychological tests are used adjunctively as part of an overall comprehensive evaluation to understand the personality, motivations, ego strengths, intelligence, defense and coping strategies, psychopathology, sexual knowledge, and sexual behaviors of the offender. The most commonly used psychological measures are the MMPI, WISC-III, or WAIS; Rorschach; Thematic Apperception Test; Bender Gestalt; and Draw-a-Person. While projective tests have been historically important, their use is limited to what they may contribute to the comprehensive assessment procedure. Neuropsychological testing and psycho-educational assessment may be required when one is suspicious of neurologically based deficits and/or learning disabilities. Measures of learning are an essential part of the assessment procedure. When indicated, family assessment measures may be administered to more fully unravel family dynamics and family process.
The following tests have been used with adolescent sex offenders:
Multiphasic Sex Inventory. The Multiphasic Sex Inventory developed by Nichols and Molinder (1984) has been standardized for adolescents. The latter measure is a 300-item, true-false test with 14 clinical and validity scales which provide measures regarding sexually offensive behavior, sexual deviations, sexual dysfunction, sexual knowledge, and sexual attitudes. There is a lack of published data on the reliability and the validity of this instrument in the adolescent population (Becker and Hunter, 1997).
Adolescent Cognitions Scale. The Adolescent Cognitions Scale for Juvenile Sex Offenders, a 32-item, forced-choice inventory, was found by Hunter et al. (1991) to have only marginal reliability and failed to discriminate between sexual and non-sexual abusers.
Adolescent Sexual Interest Card Sort. The Adolescent Sexual Interest Card Sort, a 64-item, self-report measure of sexual interest is reported to correlate poorly with phallometric assessments of sexual interest. Hunter et al. (1995) suggest caution in its use.
Child Sexual Behavior Inventory. The Child Sexual Behavior Inventory is a measure that uses parent report of sexual behavior in children aged 2 to 12 years and provides information on a number of domains to include sexual anxiety, sexual interests, sexual knowledge, sexual intrusiveness, gender-role behavior, and boundary problems (Friedrich, 1997). While increasingly used, it is limited by its reliance on parent report.
PHALLOMETRIC ASSESSMENT
Some authors have suggested the use of phallometric testing, the measuring of penile erection in response to different stimuli, as a way to determine sexual preferences (Freund, 1963). Prentky et al. (1997) noted that phallometric assessment of sexual arousal in response to depictions of children can differentiate child molesters from nonmolesrers, same-sex molesters from opposite-sex molesters, and extrafamilal molesters from incest offenders. This technique is usually reserved for the most severe and repeat sexual abusers. This procedure has generally been used with caution with minors because of the lack of empirical studies, problems of obtaining informed consent, and a reluctance to expose children and adolescents to further sexual stimulation through the portrayal of deviant sexual activities (Saunders and Awad, 1988). Becker and Hunter (1997) report that the relationship between phallometric arousal and certain clinical characteristics in the adolescent population are weaker than in the adult populati on. One study of adolescents suggests caution in the use of phallometric measures as there is an age effect determining sexual arousal and erectile measures (Kaemingk et al., 1995).
DISPOSITION
At the end of the assessment process, the clinician should be prepared to address the following issues and provide guidance to other professionals, the juvenile court, and other community agencies:
* The degree of the abuser's threat to the community, victim, or other potential victims.
* The risk of repeating the sexually aggressive behavior.
* The treatment needs of the individual and his or her family.
* The desirability of removing the abuser from his other family.
* The appropriate treatment program for the abuser (e.g., a community outpatient program) or whether he or she needs to be placed in a more restrictive environment (e.g., detention center, residential program).
TREATMENT PLANNING
There is increasing awareness that sexually abusive behavior is more than simple disorder of sexual arousal (Schwartz, 1992). There is a need for combined and integrated treatment approaches (Becker, 1994). The spectrum of emotional, behavioral, and developmental problems presented by these young people requires an integrated, multimodal treatment program which is tailored to the individual's clinical presentation and social and family support system. The predominant treatment approaches include cognitive-behavioral and psychoeducational modules, behavioral interventions, relapse prevention, psychosocial therapies, and psychopharmacological approaches (Becker, 1988, 1994; Becker and Hunter, 1993; Borduin et al., 1990; Schwartz, 1992; Shaw, 1999).
Treatment planning necessarily reflects a judgment about the level of care required. Decision-making is complicated by the heterogeneity of this population. Sexual abusers vary as to the severity of delinquency/criminality, psychopathology, sexual deviancy, the history of child maltreatment and sexual victimization, family support systems, motivation for help, and the danger they represent to the community. Hunter and Figueredo (1999) found that the degree of sexual maladjustment, denial, and the lack of a sense of accountability for one's sexual offenses predicted failure in treatment compliance in a community-based intervention program. Sexually reactive children reenacting their own sexual victimization and erorization often respond well to individual and family interventions.
More serious sexually abusive behavior is usually embedded in an array of developmental, behavioral, and emotional problems. Most juvenile sexual abusers, however, do nor meet diagnostic criteria for paraphilia nor do they have well-established patterns of deviant sexual arousal. The youthful nature of the offenders suggests that in many instances the offenders are still experimenting with different sexual activities. The plasticity of their cognitively maturing structures and the ongoing process of internalizing social and interpersonal skills suggest considerable malleability and potential for responding to appropriate social learning and treatment.
Juvenile sexual abusers usually present with considerable psychiatric comorbidity. A number of studies have documented the frequency and diversity of the psychiatric impairment (Becker et al., 1991; Kavoussi et al., 1988; Lewis et al., 1979; Shaw et al., 1993, 1996, 1999). Psychiatric diagnoses of conduct disorder, depression, anxiety, and substance abuse are common as well as evidence of character pathology and personality disorders. An understanding of the psychopathology and personality structure of juvenile sex offenders is necessary to plan a comprehensive treatment program.
An essential element in treatment planning is the evaluation of the severity of the sexual offending behavior and the risk of recurrence of sexual offending behavior. This is a difficult task even when the judgments are made by experienced clinicians (Kahn and Chambers, 1991). Since the majority of adolescent sex offenders have a history of nonsexual delinquent acts, the clinician is concerned about recidivism not only for the sexual offenses but for nonsexual delinquent acts. There is evidence that the recidivism is higher for nonsexual delinquent acts than it is for the sexual offenses.
The determination of the risk of reoffending is derived from the extensive clinical, developmental, and historical assessment of the offending youth (Gerdes et al., 1995; Rasmussen, 1999). Considerations in evaluating the risk of further sexual offenses include the frequency and diversity of the sexual offenses, the severity of the aggressive-sadistic behavior, the planfulness/impulsivity of the sexual offending behavior, prychopathology, neurological impairment, prior antisocial or violent behavior, motivation for treatment, intelligence, psychological mindedness, capacity for empathy, and family, community, and social support.
Factors to consider for placement in a more restrictive environment include the consistent need to deny sexual offenses, the lack of remorse and victim empathy, a well-established pattern of frequent and diverse committed acts of sexual aggression, the number of previous arrests, the number of victims, severity of psychopathology, failure of previous treatment efforts, the degree of compulsivity and sexual arousal, and a documented history of violent aggressive and sadistic behavior (Hunter and Figueredo, 1999; Rasmussen, 1999).
In the final stages of the clinical interview and assessment procedure, it is helpful to educate and clarify for the individual and his or her family what is going to happen and when. The role of protective services and the juvenile justice system may have to be explained. Realistic assurance should be provided that the problems are treatable, that the juvenile can learn to control and regulate sexual and aggressive impulses, and that there are specific treatment programs for children and adolescents with these problems. It should be explained to the family members what their participation in the treatment program will be. The clinician should explain his or her recommendations to the offender and should explain what is generally expected of the offender and his or her family in treatment.
Traditionally, treatability has been related to the sexual abuser's willingness to accept accountability for his or her sexual offenses manifested by (1) admission of the sexual offense, (2) acceptance of the sexual offending behavior as a problem, (3) motivation to stop sexual offending, and (4) willingness to participate fully in treatment. Other reports suggest some success with adult sexual abusers who initially denied their sexually abusive behavior (Maletzky, 1996; Schlank and Shaw, 1996). The variables that seemed to be associated with sexual recidivism are a readiness to blame the victims, the number of victims, history of sexual abuse, and the use of threats in the commission of the sexual crime (Kahn and Chambers, 1991; Rasmussen, 1999).
TREATMENT
The first task is to protect the community. It is a social imperative that the population of adolescent sex offenders who are at risk to be the adult sex offenders of tomorrow be the focus of prevention-intervention strategies. Prentky and Burgess (1990), in a cost-effective analysis of sex offender treatment programs, noted a substantial savings to the community not only in terms of cost but also in sexual reoffending behavior. Farrell and O'Brien (1988) found that the cost of providing specialized outpatient treatment was about one fifteenth of the cost that one would have to provide in an institutional or residential setting. A significant percentage of juvenile sexual abusers will respond to therapeutic intervention (Becker, 1994; Becker and Hunter, 1997; Bremer, 1992; Dwyer, 1997; Hall, 1995).
Little is known about the natural course of untreated juvenile sexual offending behavior. Dorshay (1943) found very little recidivism in a 6-year follow-up study of 256 juvenile sex offenders. He found that 7% reoffended sexually and 40% committed other, nonsexual delinquent acts. Marshall and Barbaree (1988) estimated that the recidivism rate for untreated sexual abusers is 40%. Rubinstein er al. (1993) found in a small sample of adolescent severe sex offenders that 37% had sexually reoffended and that 89% continued to commit other acts of violence.
There is general agreement that the recidivism rate for adolescent sex offenders in treatment programs is in the range of 5% to 15% (Becker, 1990; Becker and Hunter, 1993; Kahn and Chambers, 1991; Knopp, 1991; Knopp et al., 1992; Smith and Monastersky, 1986). Questions remain, in spire of the number of studies, suggesting a low recidivism rate among treated juvenile sex offenders. Critical commentaries of treatment studies have focused on methodological flaws, specifically the lack of random assignment, lack of controls, and posttest design--only paradigms (Furby et al., 1989; Miner, 1997). Maletzky (1997) suggested, however, that few would argue for untreated control groups of juvenile sex offenders. It is unethical to withhold treatment for those who want treatment and unethical to fail to protect the community by failing to intervene with juvenile sex offenders. In spite of the lack of empirical rigor, there is considerable evidence that treatment interventions are effective in interrupting the course of sexually abusive behavior. Hall (1995), in a meta-analysis of 12 studies, found a small but significant effect of treatment over no treatment.
There are several explanations as to why the adolescent offender may be more amenable to treatment than the adult sex offender:
* The adolescent offender's deviant pattern of sexual offending behavior is less deeply ingrained.
* The adolescent is still exploring alternative pathways to sexual gratification.
* The adolescent's central masturbatory fantasy is still evolving and is not fully consolidated.
* The adolescent is available for learning more effective interpersonal and social skills.
Despite the successes indicated in these studies, it is evident that not all adolescent sex offenders are equally treatable. Treat-ability depends on a number of patient characteristics and situational factors: the level of understanding of the seriousness of the offense, the motivation to discuss and understand the offense, the capacity for empathy and human relatedness, the severity of psychopathology, the entrenchment of deviant sexual arousal patterns, the type and frequency of sexual offending behavior, the aggressiveness of the sexual offense, the degree of characterological impairment, and the nature of the treatment program (Groth et al., 1981; Shaw et al., 1996; Smith and Monastersky, 1986).
The treatment of the juvenile sex offender has generally focused on a number of goals (Becker, 1994; Becker and Hunter, 1997; Ryan et al., 1987):
* Confronting the offender's denial.
* Decreasing deviant sexual arousal.
* Facilitating the development of nondeviant sexual interests.
* Promoting victim empathy.
* Enhancing social and interpersonal skills.
* Assisting with values clarification.
* Clarifying cognitive distortions.
* Teaching the juvenile to recognize the internal and external antecedents of the sexual offending behavior.
While treatment has been quite successful in reducing recidivism, adolescent sex offenders are not "cured." Treatment endeavors are organized to facilitate the sexual abuser's development of coping and adaptive strategies to prevent further sexual offenses.
The group setting is the preferred format for the treatment of sex offenders and is usually the medium through which cognitive-behavioral modalities, i.e., psychoeducational, behavioral, and relapse prevention programs, are conducted (Knopp et al., 1992; Schwartz, 1992; Smets and Cebula, 1987).
COGNITIVE-BEHAVIORAL INTERVENTIONS
Psychoeducational modules are didactic experiences that provide sexual abusers with information about sexuality, sexual deviancy, cognitive distortions, and interpersonal and social behaviors, as well as strategies for coping with aggressive and sexual impulses and anger management (Becker and Hunter, 1997; Green, 1988). This approach assumes that the offender has acquired a set of beliefs, attitudes, and expectancies which have shaped his or her sexual offending behavior and that the sexual behavior is maladaptive, contains "thinking errors," and is associated with impaired communication and social skills (Johnston and Ward, 1996). These modules are taught by a therapist who often uses workbooks and homework assignments. The setting is usually a classroom, although the intervention may take place in the context of ongoing group therapy. There is an emphasis on understanding the general patterns and determinants of sexual offending behavior, sex offender characteristics, and the spectrum of sexual offenses.
Specific Psychoeducational Modules
Victim Awareness/Empathy. The focus is on understanding the effects of sexual assault on the victim, identifying cognitive distortions and myths that support the sexual assault, and promoting participation in therapeutic endeavors.
Values Clarification. The therapist clarifies sexual values as they relate to the cessation of exploitative sexual relationships.
Cognitive Restructuring. This is an attempt to correct the cognitive distortions and the irrational beliefs that support the sexual offending behavior and to replace them with reality-focused and culturally acceptable beliefs.
Anger Management. Instruction is provided to facilitate the recognition and the development of appropriate coping strategies for managing anger.
Assertiveness Training. Training is provided to promote more appropriate self-assertive behavior to have one's needs satisfied in a reality-oriented and culturally acceptable manner.
Social Skills Training. The therapist facilitates more effective prosocial behaviors, communication skills, and interpersonal awareness.
Sexual Education. The therapist provides information regarding human sexuality, myths, sex roles, and variations of sexual behaviors.
Stress Reduction/Relaxation Management. Techniques for coping and reducing stress, anxiety, and frustration are made available to the group.
Autobiographical Awareness. Emphasis is on the individual developing an understanding of his or her own life trajectory and how the pattern of sexual offending behavior evolved over time.
Behavioral Interventions
Behavioral interventions have been used to diminish deviant sexual arousal and have been reported to be varyingly successful (Dougher, 1988b). Some of the techniques are as follows:
Covert Sensitization. In this counterconditioning paradigm, the offender learns to extinguish pleasurable responses to sexually stimulating deviant imagery through the imagining of some negative reaction or aversive stimulus. Scenes are constructed for each offender according to his or her preferred sexual-erotic fantasies (Cautela, 1966).
Assisted Covert Sensitization. Aversive stimuli such as noxious odors are used to facilitate an aversive reaction (Maletzky, 1974).
Imaginal Desensitization. The sex offender uses relaxation techniques to interrupt the sexually stimulating imagery and to inhibit the sexual arousal cycle (McConaghy et al., 1989).
Olfactory Conditioning. Sexually stimulating deviant imagery is presented which is followed by the presentation of a noxious odor.
Satiation Techniques. This involves either verbal or masturbatory satiation. The offender is first encouraged to masturbate to ejaculation in response to socially appropriate sexual fantasies with the concomitant feelings of affection and tenderness. After this experience the offender is required to masturbate to deviant sexual fantasies. If the offender becomes aroused, he or she is told to switch to an appropriate fantasy or in some intances exposed to an aversive stimulus such as ammonia (Gray, 1995). Verbal satiation requires the dictation on an audiotape of the most stimulating paraphiliac imagery for at least 30 minutes after masturbation 3 times a week. It is assumed that the paraphiliac fantasy becomes boring and subsequently extinguished (Schwartz, 1992).
Sexual Arousal Reconditioning. This involves the pairing of sexual arousal with appropriate nondeviant sexual stimulation or sexual fantasies.
Relapse Prevention
Relapse prevention was originally developed as an intervention for substance abusers but was subsequently modified for sexual abusers (Pithers et al., 1983, 1988a,b; Pithers and Gray, 1996). Ninety percent of all sex offender treatment programs in North America report using relapse prevention (Pithers and Gray, 1996). This intervention strategy assumes that sexual offenses are not capricious happenings but are the product of contextual triggers and an array of emotional and cognitive precursors. The treatment process entails the explication and definition of each phase of the sexual assault cycle, i.e., the unique characteristics of each offender's cycle so that the offender will be aware of the triggers which initiate the cycle so that he or she will be alerted and use new strategies for interrupting the sexual assault cycle (Ryan et al., 1987). Some of the emotional states that have been found to be important emotional triggers are boredom, social or sexual embarrassment, anger, fear of rejection, and numb ness (Gray and Pithers, 1993). Proulx et al. (1996) found that "negative moods and conflicts" such as anger, loneliness, and humiliation coincided with deviant sexual fantasies and increased masturbatory behavior. The goals of relapse prevention are to empower the offender to manage his or her own sexual life through a cognitive understanding of the antecedents of the sexual offending behavior and through the development of coping strategies with which to interrupt the sexual offending cycle.
PSYCHOSOCIAL INTERVENTIONS
Interpersonal therapies include traditional individual approaches, family therapy, group therapies, and the use of the therapeutic community.
Group Therapy
Group therapy with juvenile sex offenders provides a context in which the sexual abusers are unable to easily minimize, deny, or rationalize their sexual behaviors. Peer-related group therapy as the conduit for therapeutic interventions is used in a number of different ways depending on the setting (outpatient versus a more restrictive environment), group membership, the severity of the sexual offenses, group goals and objectives, whether the groups are open or closed, and the length of the group experience. Therapeutic community groups are often used in hospital or residential treatment settings as a vehicle for milieu administrative decision-making and for the monitoring of a behavioral management system. Group membership is usually either all male or all female, and very rarely is there advantage in mixing sexes in a group designed for the treatment of sex offenders.
In group therapy, the sex offender is confronted by peers who are "street-smart," who are not easily manipulated, and who are able to confront the attempts at minimization and denial. Denial may be very persistent as it is strongly supported by cognitive distortions and stereotypical views of females. The group therapy is shaped and configured by a number of therapeutic strategies which may include interpersonal, behavioral, cognitive-behavioral, psychodrama, and psychoeducational interventions.
The group is organized around an age group with some shared commonality of developmental, psychosocial, and sexual concerns. Groups are usually divided by age into early childhood, middle childhood, preadolescent, and adolescent groups. Group treatment with the younger sexually abusive child has been carried out with some success. Johnson and Berry (1989) recommended both male and female cotherapists with groups of 5 or fewer children. They are generally grouped according to age, i.e., 5 to 7, 8 to 10, and 11 to 13 years. The therapeutic intervention is structured and focused, emphasizing problemsolving skills, social skills, anger management, and cognitive strategies for controlling behavior. Thematic concerns include trust, betrayal, secrecy, guilt, labeling of affects, loss, helplessness, powerlessness, empowerment, sexual feelings, cognitive distortions, self-blame, and self-esteem. The families of these children are often seriously compromised.
An important goal in group therapy of the older juvenile offender is the achievement of a sense of group cohesion and peer acceptance. Adolescents are struggling to find their place in the peer group and to develop mutually reciprocal peer relations. They are concerned with intellectual and emotional self-expression, relationships with authority, sexual relations, moral choices, identificatory pathways, and the consolidation of a sense of self. Groups may be homogeneous or heterogeneous relative to categories of sexual offenses, open or closed, structured or unstructured. Initially, there is some advantage to the groups being structured around psychoeducational materials. The structured format provides a nidus around which the processing of emotionally arousing and threatening materials may come to the surface without being the group's being preoccupied with issues of trust (Schwartz, 1988).
The parents should receive concurrent patent counseling or be placed in a structured parent group with an emphasis on educational modules where they can discuss their own issues of victimization and more importantly focus on styles of interaction and management of their children's sexual behavior.
Individual Psychotherapy
While historically individual therapy has been a valuable intervention, it has had limited value for the individual sex offender and probably should never be relied on as the only treatment model. The advantages of individual therapy are that it provides a greater sense of confidentiality and an opportunity to develop trust in the therapeutic process. Individual therapy does provide an opportunity to assist the individual in understanding his or her psychodynamics, unresolved developmental issues, conflicts, and resistance to therapeutic intervention. Most important, individual therapy can be used to develop a therapeutic alliance which may be used to facilitate the offender's participation in other treatment modalities. There are several disadvantages to individual therapy: it is easier for the therapist to be manipulated; denial is more easily sustained; there is less therapeutic confrontation; the sexual secrets are maintained within the therapeutic dyad; and there is less opportunity to learn from others , i.e., victim empathy, offender characteristics, offense cycle, and interpersonal and social skills (Schwartz, 1988).
The therapy of juvenile sex offenders is generally characterized as one of firmness and confrontation alternating with a flexible and sympathetic stance (Muster, 1992). Confrontation is necessary to address the minimizations, denial, rationalization, and cognitive distortions which the offender presents to authority. This approach is balanced with a sensitive awareness of the offender's developmental, behavioral, and emotional problems, which not infrequently emanate from his or her own childhood history. Individual therapy is a valuable adjunct for those individuals who have been sexually abused. Frequently it is not until offenders become emotionally in contact with their own affects associated with their own history of sexual abuse and other maltreatment that they begin to extend empathy to their own victims. Individual therapy is often the treatment of choice for the younger, sexually reactive abused child who has become sexually abusive. This is particularly true for children who manifest high levels of intrapsychic conflict, emotional distress, confusion, and defensiveness around their own sexual victimization. Individual therapy allows for the development of a dyadic relationship in which issues of trust, sexually aggressive fantasies, shame, and guilt can be worked out in the context of the therapeutic relationship. In most instances, the individual therapy will be juxtaposed to other interventions to possibly include family and group therapy, cognitive-behavioral therapy, and psychopharmacological interventions.
Family Therapy
It is within the family context that many of the offender's beliefs, myths, and cognitive distortions about sexuality, aggression, and gender evolve and have been maintained. Family therapy not only provides an opportunity to understand the offender's development and coping strategies, but it provides an opportunity to correct cognitive distortions and family mythologies. Family therapy facilitates the learning of new ways of communicating and building a support system which will help interrupt the abuse cycle and ultimately be supportive to the offender's capacity for regulating and modulating sexual aggression (Schwartz, 1988; Sholevar and Schwoeri, 1999). Bischof et al. (1995) suggest that intervention strategies which have been proven effective in other delinquent groups may be effective with the families of adolescent sex offenders. Family therapy may be warranted in those instances where there is incest, especially when the sex offender remains in the family or will rejoin the nuclear family after trea tment.
PSYCHOPHARMACOLOGICAL INTERVENTIONS
Selective Serotonin Reuptake Inhibitors
The SSRIs (Greenberg and Bradford, 1997) have been shown to diminish sexual drive, sexual arousal, and sexual preoccupations. Serotonergic dysfunction has been associated with impulsivity, suicidal behavior, and aggressive behaviors (Brown et al., 1979; Coccaro et al., 1996; Linnoila et al., 1983). The SSRIs are effective in the treatment of obsessive-compulsive behaviors, and their use has been recommended for individuals with paraphiliac or nonparaphiliac compulsive sexual preoccupations. Nonparaphiliac compulsive sexual behavior (Coleman, 1992) is characterized by conventional and normative sexual behavior taken to a compulsive extreme. Others have referred to this condition as a "sexual addiction." In a retrospective study of the use of serotonin reuptake blockers in a small sample of "paraphiliac, nonparaphiliac sexual addictions and sexual obsessions," Stein et al. (1992) found those with sexual obsessions had the best response to medication. Clinical studies have suggested that fluvoxamine, clomiprami ne, and buspirone may be effective (Greenberg and Bradford, 1997).
Fluoxetine has been the agent that has been most studied, and there are a number of reports indicating that its use is associated with a reduction in paraphiliac behavior and nonparaphiliac sexual obsessions (Kafka and Prentky, 1992). Greenberg et al. (1996) carried out a retrospective study of 58 paraphiliacs treated over a 12-month period with either fluoxetine, fluvoxamine, or sertraline and found paraphiliac fantasies to be markedly reduced. A 12-week controlled retrospective study of 95 paraphiliacs treated with SSRIs matched with a control group likewise found the severity and frequency of paraphiliac fantasies significantly reduced (Greenberg and Bradford, 1997).
Antiandrogens
There has been considerable interest in the use of antiandrogen drugs to decrease sexually aggressive behavior. The evidence for the role of testosterone in sexually aggressive behavior, however, is equivocal and conflicting (Archer, 1991; Bagatell et al., 1994; Brooks and Reddon, 1996). There are reports that higher levels of testosterone are associated with increased frequency of orgasm and sexual behaviors (Knussmann et al., 1986). Other studies have failed to demonstrate increasing levels of testosterone with increased overt or covert sexual behavior (Bagatell et al., 1994). The literature is confusing because of the confound between aggressive violence and sexuality; Brooks and Reddon (1996) studied 194 adolescents 15 to 17 years of age: 75 violent offenders, 102 nonviolent offenders, and 17 sexual abusers. They found no differences between the nonviolent and sexual abusers but significantly higher levels of testosterone in the violent group. Rada et al. (1976) found no differences in testosterone level s between rapists, child molesters, and controls. When the rapists were divided in increasing gradations of violence, the most violent rapists had the highest levels of testosterone.
Bradford and Bourget (1987) reviewed the effects of antiandrogen drug treatment in sexual abusers and observed that with decreased testosterone there is a reduction in sexual drive, sexual desire, fantasy, and sexual intercourse 4 to 6 weeks after the onset of treatment. Studies have demonstrated the efficacy of medroxyprogesterone acetate (MPA) (DepoProvera(r)) in adult sex offenders (Berlin, 1989; Berlin and Meinecke, 1981). DepoProvera(r) substantially lowers serum testosterone and has been called a "sexual appetite suppressant" (Berlin, 1989).
While there is evidence that antiandrogen drug intervention decreases sexual preoccupation, sexual fantasy, and sexual behaviors, its use in adolescents has to be carefully considered (Prentky; 1997). Antiandrogen drugs such as cyproterone acetate (CPA) and MPA may delay the onset of puberty. CPA has been associated with a number of undesirable side effects such as gynecomastia, hypersomnia, fatigue, depression, and alterations in adrenal functioning, while MPA has been related to increased weight gain, gastrointestinal upset, headaches, sleep disturbances, malaise, and hyperglycemia (Katz, 1999). Luteinizing hormone-releasing hormone agonists have been shown in adults to reduce testosterone levels and concomitantly to significantly diminish deviant sexual fantasies and sexual behaviors (Katz, 1999; Rosler and Witztum, 1998).
These agents are reserved for the most severe sexual abusers and are generally discouraged for use in adolescents younger than 17 years of age. Antiandrogen medications should never be used as an exclusive treatment for paraphiliac and aggressive sexual behaviors (Prentky, 1997).
There are considerable ethical and legal implications in the use of antiandrogen medications as they are not approved by the Food and Drug Administration for the treatment of sexual abusers. A signed informed consent statement, which is comprehensive and detailed in scope with full disclosure of the side effects, risks, and potential benefits, is necessary before initiating treatment.
AFTERCARE
A sexual abuser is never cured but is rehabilitated. There is a need to provide monitoring and follow-up with continuing services. After the termination of a course of therapeutic interventions, the offender is maintained in a spectrum of continuing services which resonate with the severity of the sexual misbehaviors and psychopathology and which may include community-based outpatient treatment programs, specialized group homes, specialized foster care programs, and other specialized follow-up services.
CONFLICT OF INTEREST
As a matter of policy, some of the authors of these practice parameters are in active clinical practice and may have received income related to treatments discussed in these parameters. Some authors may be involved primarily in research or other academic endeavors and also may have received income related to treatments discussed in these parameters. To minimize the potential for these parameters to contain biased recommendations due to conflict of interest, the parameters were reviewed extensively by Work Group members, consultants, and Academy members; authors and reviewers were asked to base their recommendations on an objective evaluation of the available evidence; and authors and reviewers who believed that they might have a conflict of interest that would bias, or appear to bias, their work on these parameters were asked to notify the Academy.
SCIENTIFIC DATA AND CLINICAL CONSENSUS
Practice parameters are strategies for patient management, developed to assist clinicians in psychiatric decision-making. These parameters, based on evaluation of the scientific literature and relevant clinical consensus, describe generally accepted approaches to assess and treat specific disorders, or to perform specific medical procedures. The validity of scientific findings was judged by design, sample selection and size, inclusion of comparison groups, generalizability, and agreement with other studies. Clinical consensus was determined through extensive review by the members of the Work Group on Quality Issues, child and adolescent psychiatry consultants with expertise in the content area, the entire Academy membership, and the Academy Assembly and Council.
These parameters are not intended to define the standard of care; nor should they be deemed inclusive of all proper methods of care or exclusive of other methods of care directed at obtaining the desired results. The ultimate judgment regarding the care of a particular patient must be made by the clinician in light of all the circumstances presented by the patient and his or her family, the diagnostic and treatment options available, and available resources. Given inevitable changes in scientific information and technology, these parameters will be reviewed periodically and updated when appropriate.
REFERENCES
References marked with an asterisk are particularly recommended.
Abel G, Rouleau J, Cunningham-Rathner J (1986), Sexually aggressive behavior. In: Forensic Psychiatry and Psychology: Perspectives and Standards for Interdisciplinary Practice, Curran W, McGarry A, Shah S, eds. Philadelphia: Davis, pp 289-313
Abel GG, Becker JV, Cunningham-Rathner J, Mittelman M, Rouleau JL (1988), Multiple paraphiliac diagnoses among sex offenders. Bull Am Acad Psychiatry Law 16:153-168
Abel GG, Osborn C (1992), The paraphilias: the extent and nature of sexually deviant and criminal behavior. Psychiatr Clin North Am 15:675-687
Ageton SS (1983), Sexual Assault Among Adolescents. Lexington, MA: Lexington Books
American Academy of Child and Adolescent Psychiatry (1997), Practice parameters for the psychiatric assessment of children and adolescents. J Am Acad Child Adolesc Psychiatry 36 (suppl):4S-20S
American Psychiatric Association (1994), Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV). Washington, DC: American Psychiatric Association
Anderson PB (1996), Correlates of college women's self-reports of heterosexual aggression. Sex Abuse 8:121-131
[Araji.sup.*] SK (1997), Sexually Aggressive Children: Coming to Understand Them. Thousand Oaks, CA: Sage
Archer J (1991), The influence of testosterone on human aggression. Br J Psychol 82:1-28
Awad GA, Levene J, Saunders E (1984), A clinical study of male adolescent sex offenders. Int J Offender Ther Comp Criminol 28:105-116
Awad GA, Saunders EB (1989), Adolescent child molesters: clinical observations. Child Psychiatry Hum Dcv 19:195-206
Bagatell CJ, Heiman JR, Rivier JE, Bremner WJ (1994), Effects of endogenous testosterone and estradiol on sexual behavior in normal young men. J Clin Endocrinol Metab 78:711-716
Becker JV (1988), Adolescent sex offenders. Behav Ther 11:185-187
Becker JV (1990), Treating adolescent sexual offenders. Prof Psychol Res Pract 21:362-365
[Becker.sup.*] JV (1994), Offenders: characteristics and treatment. Future of Children: Sexual Abuse of Children 4:176-197
Becker JV, Cunningham-Rathner J, Kaplan MS (1986a), Adolescent sexual offenders: demographics, criminal, and sexual histories, and recommendations for reducing future offenses. J Interpersonal Violence 1:431-445
Becker JV, Harris CD, Sales BD (1993), Juveniles who commit sexual offenses: a critical review of research, In: Sexual Aggression, Nagayama-Hall GC, Hirschman R, Graham JR, Zaragoza T, eds. Washington, DC: Taylor and Francis, pp 215-228
[Becker.sup.*] JV, Hunter JA (1993), Aggressive sex offenders. Child Adolesc Psychiatr Clin North Am 2:477-487
Becker JV, Hunter JA (1997), Understanding and treating child and adolescent sexual offenders. In: Advances in Clinical Child Psychology. Ollendick TH, Prinz RJ, eds. New York: Plenum, pp 177-197
Becker JV, Hunter JA, Stein RM, Kaplan MS (1989), Factors associated with erection in adolescent sex offenders J Psychopathol Behav Assess 11:353-362 Becker JV, Kaplan MS, Cunningham-Rathner J, Kavoussi R (1986b), Characteristics of adolescent incest sexual perpetrators: preliminary findings. J Fam Violence 1:85-97
Becker JV, Kaplan MS, Tenke CE (1992), The relationship of abuse history, denial and erectile response profiles of adolescent sexual perpetrators. Behav Ther 23:87-97
Becker JV, Kaplan MS, Tenke CE, Tartaglini A (1991), The incidence of depressive symptomatology in juvenile sex offenders with a history of abuse. Child Abuse Negl 15:531-536
Berlin FS (1989), The paraphilia and Depo-Provera: some medical, ethical and legal considerations, Bull Am Acad Psychiatry Law 17:233-239
Berlin FS, Meinecke CF (1981), Treatment of sex offenders with antiandrogenic medication: conceptualization, review of treatment modalities, and preliminary findings. Am J Psychiatry 138:601-607
Bischof GP, Stith SM, Whitney ML (1995), Family environments of adolescent sex offenders and other juvenile delinquents. Adolescence 30:157-170
Borduin CM, Henggeler SW, Blaske DM, Stein RJ (1990), Multisystemic treatment of adolescent sexual offenders. Iut J Offender Ther Comp Criminol 34:105-113
Bradford JM, Bourger D (1987), Sexually aggressive men. Psychiatr Univ Ott 12:169-175
Bremer JF (1992), Serious juvenile sex offenders: treatment and long term follow-up. Psychiatr Ann 22:326-332
Brooks JH, Reddon JR (1996), Serum testosterone in violent and nonviolent young offenders. J Clin Psychol 52:475-483
Brown GL, Goodwin FK, Ballenger JC, Goyer PF, Major LF (1979), Aggression in humans correlates with cerebrospinal fluid amine metabolites. Psychiatr Res 1:131-139
Carpenter DR, Peed SF, Eastman B (1995), Personality characteristics of adolescent sexual offenders: a pilot study. Sex Abuse 7:195-203
Cautela JR (1966), Treatment of compulsive behavior by covert sensitization. Psychol Ret 16:33-41
Coccaro EF, Kavoussi RJ, Sheline YI, Lisk JD, Csernansky JG (1996), Impulsive aggression in personality disorder correlates with tritiated paroxetine binding in the platelet. Arch Gen Psychiatry 53:531-536
Coleman E (1990), The obsessive-compulsive model for describing compulsive sexual behavior Am J Prev Psychiatry Neural 2(3):9-14
Coleman E (1992), Is your patient suffering from compulsive sexual behavior? Psychiatr Ann 22:320-325
Cooper CL, Murphy WD, Haynes MR (1996), Characteristics of abused and nonabused adolescent sexual offenders. Sex Abuse 8:105-119
Davis GE, Leitenberg H (1987), Adolescent sex offenders. Psychol Bull 101:417-427
Davis TC, Peck GQ Storment JM (1993), Acquaintance rape and the high school student. J Adolesc Health 14:220-224
Deblinger E, McLeer SV, Arkins MS, Ralphe D, Foa E (1989), Post-traumatic stress in sexually abused, physically abused and nonabused children. Child Abuse Negl 13:403-408
Dhawan S, Marshall WL (1996), Sexual abuse histories of sexual offenders. Sex Abuse 8:7-15
Doshay LJ (1943), The Boy Sex Offender and His Later Career. New York: Grune & Stratton
Dougher MJ (1988a), Clinical assessment of the sex offenders. In: A Practitioner's Guide to Treating the Incarcerated Male Sex Offender, Schwartz BK, Cellini HR, eds. Washington, DC: US Department of Justice, National Institute of Corrections, US Government Printing Office, pp 77-84
Dougher MJ (1988b), Behavioral techniques to alter sexual arousal. In: A Practitioner's Guide to Treating the Incarcerated Male Sex Offender, Schwartz BK, Cellini HR, eds. Washington, DC: US Department of Justice, National Institute of Corrections, US Government Printing Office, pp 109-113
Dwyer SM (1997), Treatment outcome study: seventeen years after sexual offender treatment. Sex Abuse 9:149-160
Elliott DS (1994a), Youth Violence: An Overview. Boulder, CO: Center for the Study and Prevention of Violence, University of Colorado Boulder Institute for Behavioral Sciences
Elliott DS (1994b), The developmental course of sexual and nonsexual violence: results from a national longitudinal study. Paper presented at the annual meeting of the American Association for the Treatment of Sexual Abusers, San Francisco
Elliott DS (1994c), Serious violent offenders: onset, developmental course, and termination. The American Society of Criminology 1993 Presidential Address. Criminology 32:1-21
Equal Employment Opportunity Commission (1980), Title VII Guidelines on Sexual Harassment (Rules and Regulations, 74676-74677). Fed Regist 45:219